Sometimes AncestryDNA ThruLines are Wrong: Disproving an Ancestral Hint
While I’m a big fan of the hints from AncestryDNA ThruLines, we know they aren’t always correct. In my classes, I always emphasize the need to “prove or disprove” these hints.
Just yesterday, I decided to put this into practice with a ThruLines hint that resurfaced this weekend while I was teaching about AncestryDNA’s tools. ThruLines pointed to a certain couple as the potential parents of one of my direct ancestors. Hundreds of online family trees agreed. But were they right?
As it turns out, traditional research led to a strong conclusion: hundreds of trees, along with AncestryDNA ThruLines, had it wrong! This story demonstrates why we should always double-check ThruLines as well as what seems to be “accepted” family trees.
Potential Ancestor: Millie Jones
ThruLines suggests a 4th great grandmother for my father: Amelia “Millie” Frances Jones (1740-1830). This hint shows 273 DNA matches between 6 and 3,471 cM. 250 of those matches were for my dad’s direct line, but there were also suggestions through six other potential children as shown below.
Millie Jones is being suggested as a potential mother for our brick wall ancestor Sarah “Sally” Vaughan (1787-1859) who married John Bookout (1782-1857). ThruLines also suggests a potential father for Sarah: John Green Vaughan/Vaughn.
But let’s take a close look at the list of children’s names and their birth years from the above image. You might notice some interesting concerns:
-
- Thomas (born 1756)
- John (born 1765)
- Amelia (born 1770)
- William (born 1772)
- Sarah Ann (born 1774)
- Sarah (born 1787)
- William (born 1791)
Let’s examine the following:
-
- Age Gaps: There are some large age gaps in this list of children, though it might not include all of Amelia’s children.
- Late Childbearing: Amelia, the supposed mother of these children, was born around 1740. The birth years of the last two children imply that Amelia would have been around 47 and 51 at their births which seems old for the 1700s.
- Repeating Names: It’s peculiar that there are two Williams and two Sarahs, both pairs reaching adulthood and with fairly large gaps between their births.
In addition to the ThruLines hints, Ancestry has over 1200 trees that include our Sarah “Sally” Vaughan who married John Bookout. And some of these trees have a lot of sources! But none of the records I reviewed—besides those linking to some type of tree—definitively identified Sarah’s parents.
Instead of building our trees based on other people’s trees, we need to search for original records… like wills.
Searching for Wills
Since we have death dates and places for both John Vaughan and Amelia (Jones) Vaughan, I decided to search for their wills. Perhaps these wills could prove or disprove whether Sarah was their daughter.
Using a FamilySearch index for Mecklenburg County, Virginia—the place where both John and Amelia reportedly died—I was quickly able to find that both John AND Amelia had wills!
John Vaughan’s 1813 Will [1]
Hoping to find a listing for a daughter named Sarah Bookout, I turned to John’s will first. The first six bequests in his will focused on his enslaved people and items left to his wife, including “my land and plantation whereon I now live during her natural life or as long as she remains my widow…” Initially, I was concerned that he wouldn’t mention his children. However, he did, in the following way:
Children Listed in John Vaughan’s Will
-
- Stephen Vaughan’s children (inheriting their father’s portion)
- Amelia Overby
- Sarah Blank
- Mary Green
- Susannah Arrington
- John Vaughan
- William Vaughan
Interestingly, one of his daughters is named Sarah Blank. This raised a question: could there have been a clerical error where an actual blank space was mistakenly recorded as the surname “Blank”?
In 1813, my ancestor Sarah was already married to John Bookout, suggesting that John and Amelia Vaughan’s daughter Sarah, if actually married to someone named Blank, is not the same person as my ancestor.
Hoping to determine whether John’s daughter Sarah was married to a man named Blank or if it was a clerical error, I turned to Amelia’s will which was written in 1828 and proved in 1830. I hoped Amelia’s will would clearly list her children providing additional evidence of whether Sarah indeed married a Blank or a Bookout.
Amelia (Jones) Vaughan’s 1828 Will [2]
I had been thrilled to see the index listing for a will for Amelia—an uncommon find for a female in the early 1800s! As I searched for the correct image, I had one big question on my mind: would she list her children? Finding the image, I quickly scanned looking for a list of children. And, excitedly, I found it! Amelia had specificially listed her four daughters and two sons:
-
- Mary Green
- Sally Blanks
- Susannah Arrington
- Amelia Overbey
- John Vaughan
- William Vaughan
The name that immediately caught my attention was Sally Blanks. This meant that John Vaughan’s will was accurate; their daughter, Sarah “Sally,” had in fact married a man with the surname of Blank(s).
This discovery demonstrated that hundreds of trees on Ancestry, as well as the ThruLines hints, were likely wrong about the parents of Sarah Vaughan who married John Bookout! The information from these two wills was the evidence I needed to challenge the widely accepted belief that Sarah was the daughter of John and Amelia (Jones) Vaughan.
What Next? Discovering and Sharing the Truth
My next step was clear: I wanted to share the true story of this family and continue my search to identify the real parents of our Sarah “Sally” (Vaughan) Bookout. Curious to see how John and Amelia’s daughter, Sarah, was represented in FamilySearch Family Tree and WikiTree, I prepared myself for what I thought would be a lengthy task of making corrections.
FamilySearch Family Tree
I was thrilled to see that the FamilySearch agreed with my findings! The tree correctly identify John and Amelia’s daughter, Sarah Ann (1774-1859), as being married to James Blanks (1769-1852), not John Bookout. The site also includes links to both John’s and Amelia’s wills adding credibility.
WikiTree
Similarly, with contributions from the same genealogist, David Robertson, WikiTree has the same conclusion with Sarah marrying a Blanks. A listing of 60 sources along with extracts and transcriptions confirm the depth of research that has been done on this family. (David can be reached via email at davidhr@hushmail.com or at his Genealogy Facts website.)
Spreading the Word Through This Blog Post
Initially, I was prepared to correct the information on these “one world” trees, but I’m pleased to see they’ve got it right. My hope now is to spread the word through this blog post, encouraging others to update their trees accordingly. While incorrect trees often lead to misleading ThruLines, I remain optimistic. I believe that combining traditional genealogy with DNA matches will eventually unravel the mystery of my Sarah’s lineage.
Your Turn!
Have you ever encountered an AncestryDNA ThruLines hints that didn’t quite add up? I’d love to hear your story of how you uncovered the truth. Did genetic and/or traditional research help you debunk or confirm the hint? How did you share your findings with others?
Also, I’m eager to hear about those exciting moments when a ThruLines hint led you to a correct ancestor! It’s always helpful to learn from each other’s experiences in our genealogy research.
FOOTNOTES:
1. “Mecklenburg County Virginia, Will Book No. 8 with Inventories and Accounts, 1813–1819,” will for John Vaughan, written 1813, images 238–9, FamilySearch (https://familysearch.org : accessed 22 November 2023); FHL microfilm 7676177.
2. “Mecklenburg County Virginia, Will Book No. 12 with Inventories and Accounts, 1829–1832,” will for Amelia Vaughan, written 1828, image 148, FamilySearch (https://familysearch.org : accessed 22 November 2023); FHL microfilm 7676179.
I was getting a thru line for a potential parents to my 4th great grandmother Elizabeth Gresham. It was to a Dorcas Lane and Thomas Gresham. I did match descendants of Dorcas thru more than one of her children. I investigated and found that thru line had mixed up my Elizabeth with who married an Allen with Dorca’s daughter Elizabeth who married an Ellis. It took a while to figure it out.
Cathy, Thanks for sharing! And great work figuring that out!
Dana
Excellent take down!
I’m actually giving a talk at Rootstech on this very topic. It’s called Anatomy of a Hint.
Claire, That’s great! I plan on being at RootsTech and will try to check out your talk! 🙂
Dana
One of my ancestral Thru-Lines is so off, I’m surprised anyone follows the hint. My ggg grandfather, b. between 1766 and 1770 according to census records is alleged to be the son of a man born in 1760 and a woman born in 1768. I researched the couple, and they had a son with a name similar to that of my ancestor, but born in 1798.
Hi, Emily. Yikes!!! That’s pretty crazy! But I’ve definitely seen similar things in trees, so I’m betting some people would just “add” these ancestors to their trees. ((sigh))
Dana
I have posted documents to FindAGrave. It’s a search that readily comes up, making it easy for more people to find. It has helped clear up a lineage that recorded the grandfathers name instead of the fathers name on a death certificate.
Kim, That’s a great place to share your findings! I’m thankful my dad has spent a lot of time both adding family photos to Find A Grave as well as making connections between the people in our family. It’s a great site to use!
Dana
Thrulines is certainly wrong sometimes. I’d say fairly often.
My favourite example is Emma Stickley born Lovett (1838-1917). Dozens of Ancestry trees assign her to the wrong parental couple, so Thrulines takes you in the wrong direction. It may have changed lately, I haven’t looked.
It wasn’t hard to get to the truth. I had no doubt that Emma was born in Praed Street in Paddington, west London to Henry Lovett and Ann Louisa Morris. No one else agreed and assigned her to Henry Abraham Henry Lovett and Martha Elizabeth Allaway, who lived in Soho, a bit closer to the centre of London and so did Thrulines. I was confident that this was wrong. That Emma died young. My Emma’s 1857 marriage to James Stickley took place in Paddington. Her father was Henry not Abraham, although the latter might have used his middle name only. Various censuses showed her birth place as Paddington, not Soho. Finally, the 1911 census actually stated that she had been born in Praed Street, Paddington.
Case closed, surely.
So why do so many people get it wrong? I suppose they just copy each other and/or accept Ancestry’s hints uncritically. Your blog is a valuable caution.
Hi, Mick. I agree that many people – and even many of us when we first started out – just look at other people’s tree and think we’ve “solved” our research questions! It’s so easy just to add people. But we do need to be very careful. And, sadly, misinformation is rampant on both personal trees, like on Ancestry, as well as the one world trees like the FamilySearch tree and WikiTree. We just need to keep trying to correct this misinformation and make sure our own trees are the best they can be!
Thanks for sharing!
Dana
I encountered a woman who as a child (born 1837) was living with her grandparents on the 1850 census (before relationships were listed)Most Ancestry trees have her as the child of her grandparents so ThruLines are off a generation for the common ancestor correct. Later records identify her correct parents as does the extensive public surname database managed by another cousin. I contacted a known cousin who had that incorrect listing in her tree and she agreed to look into it; that was a couple months ago and she hasn’t corrected her tree yet. Most other times I have tried to get people to change their trees has been unsuccessful so I haven’t gone on a campaign to get it fixed – yet. I did add notes to the person in my tree in case anyone bothers to look. Sometimes, they do.
There was a time when ThruLInes would use what we have in our tree over others’ trees. In this case, it did not. I have the match and her correct line in my tree and connected to the DNA match, yet ThruLines is still pulling from other incorrect trees.
Thank you for this article and for all you do to help us 🙂
Hi, Barbara. Thanks for sharing! I’ve also got several instances of grandchildren living with their grandparents and people thinking they’ve located the parents. It’s quite possible we’ve done the same thing! Sometimes it’s hard to figure it out. That’s also one example of why we really shouldn’t rely on just one record, right?
Best wishes!
Dana
I had a bit of a different problem: My husband’s 2nd great grandfather was named Lorenzo Childs. Born in either Vermont or Massachusetts, died in PA 1864. I entered all the info I had for him. Then I searched and found a Lorenzo Whitney Childs, born in Vermont about the same time, died in Indiana 1901, but buried in Vermont. Married in Vermont. I added him as a floating person thinking I could maybe match them. ThruLines then put him in the position of hubby’s 2nd ggf. Many many people have copied that on their trees, and no doubt it’s in their ThruLines, as well as family members. No one wants to give up on Lorenzo Whitney Childs because there is much more info on him. At least ThruLines corrected it on hubby’s tree.
Genie, Yikes! Dealing with two men (or women) of the same name can be quite difficult. And I’m guessing everyone would rather have the person with a lot of records than the one who is difficult to research! So it’s easy to add them to our trees. Ugh! Maybe one lesson is to take it slow and steady and really research people before we add them to our trees.
Sincerely,
Dana
I have this same situation with Willis Senter/Center.
My Willis Center was b. 1786 in NC, moved to SC, and walked to KY with his wife and sister-in-law, dying in Allen County, KY in 1858. It was the bio of the sister-in-law’s grandson William Brown in a different county that verified the information.
95% of the descendants of Willis Center, attach Willis S. Senter as their ancestor. The second Willis Senter b. 1785 d. 1845 Roane Co, TN had proven parents. My Willis has minimal paper trails, but DNA connects to Rev. war soldier. William Senter/Sentell b. VA d. NC.
Then there are the two separate Christopher Haines who served in the Rev. War from VA and died in 1846, in Allen County, KY. People confuse him with Christopher Haynes of Jefferson Co, TN who served in the TN House of Representatives in the early 1800s. The fact that the 1st Christopher’s brother lived in the same county as the 2nd Christopher doesn’t help the situation. However, deeds in Jefferson Co, TN proved they were not the same men. How could the same person live in two different counties in two different states at the same time in the early 1800s? The first left multiple descendants, the 2nd was never married and left no descendants. Research is complicated because early censuses in eastern TN are missing, and the Allen County, KY courthouse burned in 1902 taking most of the records.
Another ancestor with questionable Thrulines attachments is Enoch Berry, b. 1793 VA d. 1834 Allen County, KY. His father was Rev. War soldier Augustine Berry b. 1761 VA, d. 1811 Barren County, KY. (Allen County was formed from part of Barren in 1815). People want to attach Augustine as a child to Enoch Berry 1700-1763 of King George Co, VA. This Enoch married Drucella Bunbury in 1726. The odds of them having a child in 1761 is not probable.
These is just 3 of many situations where ThruLines is wrong. I’ve given up after 30 years trying to correct errors.
Hi, Sharon. It sounds like you’ve been thorough pursing many of these lines and seeing how they’re wrong. But hopefully some will lead in the right direction, too! I also just love to cluster my matches and determine a common ancestor that way. But, I like that these ThruLines hints are quick and give me a potential ancestor to prove or disprove.
Thanks for sharing!
Dana
My ancestor John Price Sr of Shenandoah County, Virginia has been confused with many men of the same name. His will is recorded in Will Book B Page 95 Shenandoah County. He names his wife Sarah, five sons: Edward, John, Thomas, Sampson and Zachariah and a young daughter Nancy. A 1796 Deed for land willed to son Edward identified four additional daughters plus spouses of all the living siblings. Shenandoah County Deed Book K Page 425. Edward died without issue. Edward’s property was then sold by the surviving brothers and sisters who are identified as such and also as children of John Price, deceased. Additional documentation for sons Thomas and Sampson was found in their Revolutionary War Pension Applications. One incorrect John Price also had a son named Thomas. However, that Thomas lived his entire life in Hanover County according to his Revolutionary War Pension Application. The third John Price and his children were identified in a will, Essex County Will Book 12, Pages 448-449 and Chancery Records of Cumberland County found online at https://www.lva.virginia.gov/chancery/. A fourth John Price who also had a wife named Sarah died in Halifax County, Virginia in 1772. His children were also identified in a will, Halifax County Will Book O Page 323. What is frustrating is those who do not pay attention to detail or add incorrect information in hopes of extending the line or connecting to prominent individuals. Additional proof was found using YDNA results as those connecting to John Price of Shenandoah County are in a different haplogroup except the John Price of Halifax County. I have posted information for this family on family search and wiki tree.
Darlene, What great documents piecing this family together! And good for you for posting at both FamilySearch and WikiTree. I know it’s frustrating to see people confusing men of the same name. It sounds like it would be a great case for the NGSQ if you wrote it up. 😉 And, of course the Y-DNA is a great addition to your proof!
Best wishes & wish we shared family and could collaborate! 😉
Dana
When I use thru lines to add/link DNA matches to my trees I always independently research the older relationships. I will accept people’s trees back to grandparents but everything past that I research by myself back to original records.
Sarah, If there is information there, I will try to prove those last few links, too. But, yes, I sometimes just accept their parents if all the other pieces are falling in place!
Thanks for sharing,
Dana
My 3x great grandfather James McLean was born in Ireland in 1795. He married fellow immigrant Sarah McLain in Philadelphia and they settled in Pittsburgh, Pennsylvania. Although the surname appears as Maclean as middle names in later generations, all kinds of permutations of their last names appear in records. Of the 7 children listed in James 1854 will, only two daughters have descendants, my 2x great grandmother Susanna Shields and her sister Sarah Ann Munn. Thrulines did correctly direct me to descendants of Sarah Ann, despite permutations of the surnames in their trees.. However, first someone someone decided my family in the 1850 census of Pittsburgh must be their Sarah McClain (who census records indicate was born in Maine) passed to her Indiana state born Susannah who married Levi Miller in 1845, and passed the album on down the Miller line. They attached beautiful pictures of unknown people to my family. The 1845 marriage indicated their Susannah was not my Yet to be married Susanna. Her brother William McLean died at age 14 so could not be the 30s something man in the picture attributed to him, etc. I had contacted someone in that line to explain the error years ago. But was dismayed last week to discover Sarah Ann’s descendants are still passing on these photos on their trees. Second Sarah’s descendants believe an Indiana state born John McClain who married an Ireland born Margaret McGowen in New York are the parents of my Ireland born James McLean. Unlikely to go back to Ireland to have a son. But need to research those people further to disprove this.
Hi, Mary. That’s frustrating! I wonder how many other “wrong” photos are being shared on trees. It’s so exciting to see a photo, I’m sure many people just grab them and share! I hope you’re able to straighten this out, but usually we can only fix our own trees. And it sure is interesting how people will just ‘ignore’ the parts of a person who don’t line up with their person.
Thanks for sharing,
Dana
FamilySearch has been a goldmine for finding those old documents like wills.
With my Virginia ancestors, I also find the Virginia State Library to be great.
https://www.virginiamemory.com/collections/chancery/
Their collection of Chancery records helped me confirm an ancestral connection.
Great post!
Hi, Marti. Yes, FamilySearch is a WONDERFUL source!!! And thanks for sharing about Virginia! And thanks for sharing about the chancery records on Virginia Memory!
Dana
FamilySearch has been a goldmine for finding those old documents like wills.
With my Virginia ancestors, I also find the Virginia State Library to be a great resource.
https://www.virginiamemory.com/collections/chancery/
Their collection of Chancery records helped me confirm an ancestral connection.
Great post!
Thank you for this blog post! I know I sometimes fill in slots on my Ancestry family tree in a speculative fashion, trying to actually generate hints that might clear some of my own brick walls. But I am still always looking for some verification beyond just someone else’s family tree data. I didn’t realize though that enough repetition of speculative ancestry could influence the ThruLines.
On another note… I wonder if our trees might cross somewhere as I have ancestors in that part of Virginia. Does your John Green Vaughn have Green/Greene ancestors?
Hi, Erin. Yes, sadly our speculations can both 1) be copied by others and 2) affect ThruLines (probably even once we’ve taken that potential ancestor away). I’m working on Part 2 of this post and I think it’s interesting to see “why” this ThruLines formed.
And I don’t believe I’m actually related to John Green Vaughan, but maybe we connect some other way. 🙂
Dana
Pingback: Friday’s Family History Finds | Empty Branches on the Family Tree
I just got my first DNA results the other day. I am being shown results that seem to be DNA-proven but don’t make sense — I’ll see multiple matches to a known ancestor’s potential uncles or cousins, but the people shown as the parents were too young or too old or something to be the parents. For example, it’s eager to suggest that Donald and Daisy are the parents of Huey, Dewey and Louie, and that they’ve found a father for Donald named “Uncle Scrooge McDuck.” Is there a way to get behind the proposed parents and see what matches connect me to THEIR siblings on each side, WITHOUT accepting the faulty suggestion that they are my ancestors?
Hi, John. How exciting to see new DNA results! I’m assuming you are talking about ThruLines, and yes, they can often be way off base. I’m not 100% sure what you’re asking. Can you read the new post I did today on ThruLines and see if that helps? If not, please ask again!
Dana
Pingback: AncestryDNA ThruLines Part 2: Evaluating Matches and Exploring Y-DNA Possibilities - Dana Leeds